The invention is directed to a system for efficiently processing information originating from hard copy documents. More specifically, the invention is directed to a hard copy document to application program interface which minimizes the need to manually process hard copy documents.
In the past, information contained on hard copy documents was manually entered into a computer via the input controller of a particular computer. The original document was then filed away for future reference. Automatic input of data was limited to the input of Magnetic Ink Character Recognition (MICR) data and to Optical Character Recognition (OCR) data. This fixed-position data was forwarded directly to a dedicated computer application specifically designed to accommodate the input format. In more recent years typewritten text has been mechanically inputted into a computer via a text file. Examples of this latter type of system are word processors and photo-typesetters.
These conventional systems have limitations which decrease the efficiency of processing information from a hard copy document. For example, the systems discussed above are limited in their application to MICR, OCR, or typewritten data. Parsing and processing data is limited to the particular requirements of the particular computer application which requires the input data. In addition, in these conventional systems, the actual hard copy document must be retained for future reference at great expense.
In a sophisticated computer network, different users may require different portions of the information contained on a hard copy document. For example, if the hard copy document is an invoice returned with payment of a bill, the accounting department may need all of the monetary information contained on the bill while the mailroom may need only customer address information, to update a customer's address. Therefore, there is a need for a system in which specific information from a hard copy document can be selectively distributed to various users.
Another problem with conventional systems is that users, even within the same company, may require that the information extracted from a hard copy document be transmitted to a particular application program in a specific transmission format. For example, one department in a company may use a particular application program which must receive information using a particular character as a delimiter and other departments may require the information in a different format using different delimiters.
Another problem, particularly for small businesses, is that current systems can not efficiently accommodate the inputting of information from a diversity of hard copy documents. A large business which receives many forms in the same format can afford a system which inputs a high volume of information in that format into memory. For example, it is cost-effective for a bank which processes hundreds of thousands of checks a month to buy a dedicated machine which can read information off of checks having a rigidly defined, or fixed, format. However, as the diversity of forms received by a business increases relative to the number of forms that must be processed, it becomes less cost-effective to design a dedicated machine for processing each type of form format. This problem is particularly significant in small businesses which may, for example, receive fifty invoices a month, all in different, non-fixed, formats. It is frequently not cost-effective for a small business to design dedicated systems for inputting information in each of these various formats. This leaves a small business with no other practical alternative than to manually input the information off of each invoice each month.